Within the overarching theme of "New Approaches to Heterogeneity in Dementia," Project 2 has focused on Alzheimer's disease (AD). The insights we gained during the preceding funding period and the ever increasing threat AD poses to public health have motivated us to maintain this focus in the current proposal. We will also continue to utilize genetically engineered mice with neuronal expression of human amyloid precursor proteins (hAPP) and amyloid-p (AP) peptides, because there is substantial evidence for mechanistically informative overlap between these models and the human condition. In our original application, we promised to shed light on the processes by which Ap elicits neuronal deficits. We found that neurons in the dentate gyrus and entorhinal cortex - brain regions affected early and severely by AD - are particularly vulnerable to the Ap-induced depletion of proteins that are critical for learning and memory. Several molecules were identified that may mediate this process. We also identified strategies to prevent Apinduced neuronal deficits in hAPP mice. Our new proposal builds on the most promising findings we obtained during the preceding funding period. Specifically, we discovered that the depletion of calciumdependent proteins and associated memory deficits in hAPP mice are likely caused by spontaneous nonconvulsive epileptiform activity in cortical and hippocampal networks. Memory deficits, depletions of calciumdependent proteins, and abnormal network activity could be prevented in hAPP mice through a genetic manipulation that blocks neuronal overexcitation. Independent lines of evidence suggest that epileptiform activity may also play a pathogenic role in humans with AD. We therefore postulate that aberrant excitatory neuronal activity might play an important causal role in the pathogenesis of Ap-induced cognitive impairments in hAPP mice and in AD. This hypothesis will be tested in three new specific aims. In Aim 1, we will examine whether markers of abnormal neuronal activity are increased in brains of" patients with mild cognitive impairment (MCI), AD, or other dementias. In Aim 2, we will test whether available anti-epileptic drugs can prevent or reverse EEC abnormalities in AD-related mouse models. In Aim 3, we will test whether any of these anti-epileptic drugs can also prevent or reverse cognitive deficits in these models. Confirmation of these untested hypotheses should help elucidate the mechanisms that underlie Ap-dependent cognitive deficits and pave the way for the development of better treatments for AD. Although there is plenty of ' evidence for a potential role of epilepsy in the development of AD, there appear to have been no rigorous clinical trials of anti-epileptic drugs in patients with MCI or early AD. The experiments described in our application could pave the path towards such a clinical trial and provide critical guidance in the selection of the most promising drugs. The proposed ADRC will provide an ideal environment for us to achieve these goals.